PROJECT SUMMARY/ABSTRACT BACKGROUND: Several independent lines of evidence suggest that sleep disturbance may directly contribute to the generation and maintenance of neuropsychiatric symptoms (NPS) including anxiety, depression, agitation, irritability, and apathy through fronto-limbic brain networks that regulate emotion, particularly in regions of the prefrontal cortex (PFC) and limbic areas (amygdala). However, the model of sleep disturbance contributing to increased NPS through impairments in fronto-limbic function has not yet been tested in a sample of participants with or at high-risk for developing Alzheimer?s Disease (AD). OBJECTIVE: We aim to test this model in patients with mild cognitive impairment (MCI) and mild AD by characterizing associations between sleep disruption, fronto-limbic function while regulating emotion, and NPS at baseline and by experimentally manipulating sleep to determine whether changes in sleep cause downstream alterations in fronto-limbic functioning and NPS. DESIGN/METHODS: Our hypotheses will be tested in a 2-arm randomized controlled mechanistic trial with 150 patients with MCI and mild AD experiencing sleep disturbances who are also experiencing emotional distress and other behavioral symptoms. Participants will be randomized in a 1:1 ratio to receive a sleep manipulation (Cognitive Behavioral Therapy for Insomnia; CBT-I) or to the credible control treatment for insomnia group. Both interventions will be administered in six sessions delivered over eight weeks. CBT-I is an efficacious behavioral intervention specifically targeted at improving sleep patterns through a combination of sleep restriction, stimulus control, cognitive therapy targeting dysfunctional beliefs about sleep, and sleep hygiene education. At baseline, and post-sleep manipulation we will assay each participant?s fronto-limbic functioning while regulating emotions (amygdala reactivity and PFC-amygdala connectivity), NPS using the Neuropsychiatric Inventory (NPI), and sleep efficiency using high-density EEG collected overnight during sleep. SPECIFIC AIMS are to 1) characterize baseline associations among sleep disturbances, emotion regulation brain function, and NPS prior to a sleep manipulation, 2) test that sleep-induced fronto-limbic brain function improvement mediates the association between sleep improvement and NPS reductions, and 3) determine baseline predictors of the NPS improvement. NOVELTY & IMPACT: Using a multi-methodological approach within a mechanistic trial framework by causally manipulating sleep, we will uncover potential NPS mechanisms across multiple units of analysis (brain circuit, physiological, behavioral, and self-report). Our results will advance a mechanistic understanding of how sleep disturbances and fronto-limbic brain function while regulating emotions may underlie the emotionally distressing and economically relevant problem of NPS in early AD patients. These results would be a necessary first step in the development of sleep based, mechanism-focused preventative strategies and treatments that are more personalized for the individual.